a5huynh
Joined 19 years ago
Karma
202
Submissions
66
Member since
Mar 9
Recent submissions
Comment · 1 year ago
Side-note, if anyone wants to really dig into all the data available about bills (including votes, attachments, etc.), this is a great place to start: https://github.com/unitedstates/congress
There's excellent documentation on the formats and how to access all the data.
Comment · 2 years ago
Not op but one my favorites is: https://www.youtube.com/@Settledrs
Very entertaining to watch and explains things so even non players can understand the sheer absurdity of some of the attempts.
Comment · 2 years ago
I don't think its too misleading because in both cases that's how many vehicles are affected at very moment by the recall.
Toyota shipped/sold more as mentioned in the article, but those are unaffected. Likewise with the Tesla, any shipped afterward the defect was discovered are unaffected.
Comment · 2 years ago
Just to add context since I was curious about "Toyota had it wheels literally falling off".
That was a recall from 2022 (https://www.cnn.com/2022/10/06/business/toyota-bz4x-wheel-fi...) for 260 vehicles (their model BZ4X electric SUV).
The Cybertruck recall affects 3,878 vehicles (https://www.caranddriver.com/news/a60538687/2024-tesla-cyber...).
Comment · 2 years ago
This reminds me of Floneum (https://github.com/floneum/floneum), this open-sourced tool for graph-based workflows using local LLMs.
More for personal use and not quite as polished but a decent alternative for those looking to play around with the idea locally.
Comment · 2 years ago
I've been using a combo of LLMs + live transcription to build a passive assistant that keeps track of talking points and can pull out data/tasks from a conversation you're having (https://sightglass.ai or here's a demo of me using it: https://www.loom.com/share/0220ca03bce341669d314d4254872226)
So far this is being used for:
- Sales -> guiding new recruits during more complex client calls
- HR -> Capturing respones during screening interviews
If you'd like to try this out feel free to DM me or email me at andrew at sightglass.ai, we're looking for more testers!
Comment · 3 years ago
It's in San Francisco. It was announced Sept. 6th on their blog: https://openai.com/blog/announcing-openai-devday
Comment · 3 years ago
A problem with fine tuning based on organization data is that if the underlying data changes, you'd need to fine-tune the model again each change. This might be okay for one-off changes (such as the name of the model in the example) but if it costs $300 each time (not to mention the time spent) and you have 100s/1,000s of changes per month, it's not really viable.
Comment · 3 years ago
If you want to play around with Stable Diffusion XL: https://clipdrop.co
Comment · 3 years ago
Hey HN!
We've been building a semantic search engine for audio content focusing on podcasts. We're getting an early version for feedback.
You can search or ask questions about a particular podcast episode or feed and get back an answer as well as link to the relevant podcasts.
We also let you follow your favorite podcasts and receive summaries in your inbox whenever a new episode comes out
Comment · 3 years ago
If you're running the text-generation-webui (https://github.com/oobabooga/text-generation-webui) it has the ability to train LoRAs.
It'll require a beefy GPU but I've seen some fun examples like someone training a LoRA on Skyrim books.
Comment · 3 years ago
I recently found this list of models that works with llama.cpp: https://rentry.org/nur779 (with dl links, albeit given llama's licensing gray area, use at your own risk)
The latest so far would be Vicuna, whose weights were just recently release.
Comment · 3 years ago
As an alternative for purely local LLMs, I've been having fun with this setup: https://github.com/oobabooga/text-generation-webui
Comment · 3 years ago
Note that what they released are the delta weights from the og LLaMa model. To play around with it, you'll need to grab the original LLaMA 13B model and apply the changes.
> We release Vicuna weights as delta weights to comply with the LLaMA model
> license. You can add our delta to the original LLaMA weights to obtain
> the Vicuna weights.
Currently:
* Building agentic commerce
Previously:
* data & AI @ customer support startup
* founder, AI agents for eDiscovery startup (raised pre-seed)
* founding team, chief data science guy @ insurance startup (made it to Series A before being acquired).
* manager & turning coffee into code @ NLP machine learning startup).
* founder, CTO @ sustainability startup (raised seed money)
Before That:
* Machine Learning Ph.D Student @ UC San Diego.
* Graduate Student Researcher on a National Geographic Expedition.
* Stripe CTF Finalist
* 3 time Yahoo! Hackathon winner. Global Hackathon Finalist.
[ my public key: https://keybase.io/a5huynh; my proof: https://keybase.io/a5huynh/sigs/Df-98a5Ya5JkscNzgBwy5xxR-P6vhb8Rf6aM4zhr7y4 ]